本文利用Jorda(2005)的局部線性投影(local linear projections)之衝擊反應函數(impulse response functions)估計法來探討台灣、日本及韓國等三個東北亞國家實質匯率的半衰期(half-life)。同時,亦採用Amadi(2004)的方法來計算戀家特性(home bias)的指標,以納入實證模型中當作控制因素。實證結果發現:在加入戀家行為因素後,使實質匯率的半衰期相對下降許多。因此,在結合適當的實證方法及戀家行為因素下,本文之實證結果有助於解釋對於實質匯率調整速度太慢的“購買力平價之謎”(Purchasing Power Parity Puzzle),即過去文獻得到的過長之實質匯率半衰期可能是因忽略了戀家行為的因素。
This paper investigates the half-life of real exchange rates for three Northeast Asian countries after taking into account the impact of home bias using Jorda's (2005) methodology of local linear projections to compute the median unbiased estimation. The index of home bias is computed based on Amadi's (2004) formula. Our results indicate that the half-life of real exchange rates decreases if home bias is controlled. Thus, employing home bias index together with proper estimation technique can help reduce the half-life of real exchange rates.